Structural representatives, selection

Penicillin G acylase (PGA, EC 3.5.1.11, penicillin G amidase) catalyzes the hydrolysis of the phenylacetyl side chain of penicillin to give 6-aminopenicillanic acid. PGA accepts only phenylacetyl and structurally similar groups (phenoxyacetyl, 4-pyridylacetyl) in the acyl moiety of the substrates, whereas a wide range of structures are tolerated in the amine part [100]. A representative selection of amide substrates, which have been hydrolyzed in a highly selective fashion, is depicted in Figure 6.36. 

A comprehensive list of P-gp modulators or inhibitors, classified according to their chemical structures, has been published recently [87]. This shows that the structures of inhibitors are almost as heterogeneous as those of the substrates. A small but representative selection of inhibitors is shown in Fig. 20.12 and Table 20.1. In an attempt to clarify the different mechanisms of P-gp modulation or inhibition, the H-bonding concept discussed above is applied. To this end, the modulators or inhibitors in Table 20.1 were ordered according to their H-bond acceptor potential and divided in three groups comprising compounds with (i) a low EUh (<2 i.e., not transported) (ii) an intermediate EUh (— 3—6) and (iii) a high H-bond acceptor potential ( EUh > 10 i.e., transported slowly). 

Commercially viable systems for the decolorisation of spent dyebaths can be based on hydrogen peroxide treatment initiated by UV radiation. A representative selection of six disulphonated monoazo acid dyes and two disazo disulphonated types was exposed for various times in a pilot-scale photochemical reactor of this kind. The controlling parameters were dye structure, pH, peroxide dosage and UV light intensity [39]. In a wider survey of the response of various classes of dyes to the combination of UV radiation and hydrogen peroxide, viable candidates for further in-plant treatment trials were the water-soluble reactive, direct, acid and basic classes. On the other hand, water-insoluble colorants such as disperse and vat dyes did not appear to be viable [40]. 

Throughout this review, emphasis will be placed on publications in scientific journals patents claiming biological activities will be mentioned only briefly. The formulae given for classes of compounds represent selected typical structures. In order to ensure lucidity of the presentation, not all the substituents listed in a patent are shown. 

The biosynthesis and endocrine regulation of pheromone production in beetles has been reviewed [33, 34]. Nevertheless, some more general pathways will be briefly discussed here. As corresponding structures are widespread among insects [2], the examples shown here are selected mostly from taxa other than beetles. Structures representing beetle pheromones will be shown in the context of the discussion of the corresponding species. 

At this point it may be valuable to digress a moment and discuss the state-of-the-knowledge in the field of Fe-S proteins by the ntid-1970 s. At this time there were three known structures found in nature, IFe as represented by rubredoxin, the [2Fe-2S] cluster as represented by plant ferredoxins, and the [4Fe-4S] cluster as found in many bacterial ferredoxins (24). The schematic structures and selected properties are listed in Table I. 

The coordination number for these complexes may be six or seven and extensive X-ray structural studies have been carried out, particularly on the seven-coordinate complexes. The geometry and some other details of a representative selection of these compounds are shown in Table 5. In some cases, e.g. [Mo(CO)2Cl2(dpam)2], one potentially chelating ligand is... 

The above predictions, that the stereochemistry is largely controlled by the normalized bite of the bidentate ligands, is in general agreement with the known structures, a representative selection being given in Table 19. 

In the selection of a screening system, the determination of the set of structural characteristics to act as the screens is a major problem. A proper balance must be established between the cost of generating, storing, and searching the screens, and insuring that the searches at the screen level achieve complete recall. In addition, the structural characteristics selected as screens should occur with a distribution as even as possible. Because of the uneven distribution of structural characteristics, this represents a significant problem. 

The discussion above indicates that QPPR models must be selected carefully, considering the structure of query compound and its relationship to the structures represented in the training set. It is often useful to employ different models and to compare the results. 

---